Ask HN: Downloadable or SAAS?
2 comments
I would go SaaS only. The ones who insist on having everything on their servers are probably going to be high-maintenance customers you don't want to deal with anyway. Remember the 80/20 rule. The 20% of your clients who want the download will take up 80% of your time.
You have to convince them that SaaS is the better deal - regular, automatic updates, tight security, redundant backups, and reliable access - and they won't want to consider the downloadable version.
You should prefer SaaS because you can charge them more in the long run if you get them dependent on you, in addition to reducing the setup time and effort.
You have to convince them that SaaS is the better deal - regular, automatic updates, tight security, redundant backups, and reliable access - and they won't want to consider the downloadable version.
You should prefer SaaS because you can charge them more in the long run if you get them dependent on you, in addition to reducing the setup time and effort.
I feel like you are fooling yourself here. Many small companies want to download and install not because they are difficult customers, but because they are either a) operating on a limited budget or b) risk adverse.
As you say yourself, getting them to sign up for SaaS is better for the developer because you can charge them more long term. By the same token, many customers just want a simple solution that works, without additional costs.
If I am a small business looking to invest in some software, why would I ever go with a SaaS from an unestablished company just starting out? As anyone here on yc should know, -most- companies fail. Odds are if I sign up for SaaS from an unestablished company (which the original poster's company would be if he is just starting up), they could very well go under in the future. Why would I open -my- small business to that risk if there is a competing program available I could download and install and no longer worry about whether the service provider will be around? That isn't being difficult, that's just being responsible.
As I say in another comment, I would probably still start with SaaS, just because it is easier on the developer starting out. Less worries re:installation process, fewer platforms to support, etc. But I just disagree completely with your characterization of customers who prefer a downloadable product.
As you say yourself, getting them to sign up for SaaS is better for the developer because you can charge them more long term. By the same token, many customers just want a simple solution that works, without additional costs.
If I am a small business looking to invest in some software, why would I ever go with a SaaS from an unestablished company just starting out? As anyone here on yc should know, -most- companies fail. Odds are if I sign up for SaaS from an unestablished company (which the original poster's company would be if he is just starting up), they could very well go under in the future. Why would I open -my- small business to that risk if there is a competing program available I could download and install and no longer worry about whether the service provider will be around? That isn't being difficult, that's just being responsible.
As I say in another comment, I would probably still start with SaaS, just because it is easier on the developer starting out. Less worries re:installation process, fewer platforms to support, etc. But I just disagree completely with your characterization of customers who prefer a downloadable product.
I agree with this, but would add one point. You can add a downloadable option later if it is something your customers say they want.
IMO, SaaS is good for the developer (of the SaaS), and for the (very) small business customer. Downloadable is better for the slightly larger business, or a business that has more technical know-how than software budget.
It follows that since you are the developer, and since your are starting out, you should favor the method of deployment which favors -you- and then expand on that as needed/requested. To me that means start out SaaS, and expand your offering as required.
I disagree with the poster who says that the customers who want downloadable will be high-maintenance. There may be some, but really the customers who want downloadable are simply more risk-adverse. We install in-house only currently (not my decision, but policy nevertheless), and beyond one installation case, I don't know that we've logged any additional support over 5 years, across multiple applications. It is just more work for the developer up front to ensure that your installation process supports all possible configurations you will be installing to.
It follows that since you are the developer, and since your are starting out, you should favor the method of deployment which favors -you- and then expand on that as needed/requested. To me that means start out SaaS, and expand your offering as required.
I disagree with the poster who says that the customers who want downloadable will be high-maintenance. There may be some, but really the customers who want downloadable are simply more risk-adverse. We install in-house only currently (not my decision, but policy nevertheless), and beyond one installation case, I don't know that we've logged any additional support over 5 years, across multiple applications. It is just more work for the developer up front to ensure that your installation process supports all possible configurations you will be installing to.
From what I've seen, there's a large number of businesses that are only willing to rely on software that they can install and monitor on their own servers. Many have policies where they simply won't use a software product that resides (along with their data) on someone else's servers.
While I am going to try and make it as attractive as possible to use my product as a web service (which will include making all of their data fully exportable in an open format), should I start out offering the product as both a web service AND a downloadable product? Should I start out as JUST a web service and gauge customer reaction?
The only reason I would prefer to offer the product as a web service only is that it will reduce the amount of time I would have to spend maintaining the downloadable version, and eliminate support and development issues related to it being installed on third party servers running god-only-knows what. I'd rather funnel that time and effort into the web service. Piracy is not necessarily among my list of concerns.
What are everyone's thoughts?